A novel three-way rearrangement involving ETV6 (12p13) and ABL1 (9q34) with an unknown partner on 3p25 resulting in a possible ETV6-ABL1 fusion in a patient with acute myeloid leukemia: a case report and a review of the literature

This case highlights the formation of a potential ETV6-ABL1 fusion gene as a result of a complex, three-way rearrangement. The conventional and molecular cytogenetic findings in this case suggest an insertion of ETV6 in the short arm of the derivative chromosome 3 followed by a reciprocal translocation involving the same derivative chromosome 3 and ABL1 (9q34), resulting in the potential fusion gene. The breakpoint on chromosome 3 at which the aforementioned rearrangements occurred - 3p25 - harbors ANKRD28 (3p25.1), which has been implicated in AML in the context of t(3;11)(p25;p15) involving NUP98 (11p15) [10]. In other studies, 3p25 was found to be the most frequently deleted chromosomal band on 3p in AML [11]. Given this information, the involvement of 3p25 in this three-way rearrangement may result in deregulation of particular target genes relevant to leukemogenesis in this region. The translocation that occurred concomitant to the three-way rearrangement - t(5;18)(q13;p11.2) - has not been observed in AML and has not been associated with any clinical or hematopathologic features.

To date, five cases of AML bearing a possible ETV6-ABL1 fusion have been reported (Table 1). Two of the cases were designated M1, two were M6, and one was not reported. One of the cases showed a straightforward t(9;12)(q34;p13) with concurrent FISH studies showing the 3’ ABL1 signal on the 12p, albeit in the context of a complex karyotype [4]. Two of the cases showed rearrangements involving a third chromosome - Golub et al. reported a case of AML (M6) with a t(9;12;14)(q34;p13;q22) without additional FISH studies, and La Starza et al. reported a case of AML (M1) with a t(8;12)(p21;p13) that showed colocalization of 5’ ETV6 and 3’ ABL1 signals on 8p21 by FISH [3, 4]. The two remaining cases showed normal karyotypes by conventional cytogenetics, but showed 3’ ABL1 signals on 12p [13, 14]. Among all of these cases, reported secondary abnormalities observed included gains of chromosomes 8, 9, 11, 12, 14, 17, and 19, as well as a t(1;22) [4, 14].

The ETV6-ABL1 fusion includes a helix loop helix (HLH) domain of ETV6 and tyrosine kinase domain of ABL1, and each domain is necessary for constitutive phosphorylation to occur [4]. Millon et al. found that mice transplanted with ETV6-ABL1-positive hematopoietic stem cells developed CML-like myeloproliferative disease, and that the TEL pointed homology oligomerization domain was essential to ETV6-ABL1-driven leukemogenesis [15]. It is well known that the ABL1 kinase has altered catalytic specificity in human leukemia. Co-immunoprecipitation studies show that the ETV6-ABL1 fusion protein tends to form complexes with CrkL in Ba/F3 cells, and this interaction phosphorylates CrkL and possibly CrkII [16]. However, it is known that the in vitro studies tend to have a wider range of substrates than the cellular forms [16]. Further analysis of Crk and CrkL adaptor proteins show that they play an essential role in integrating signals from a wide variety of sources such as apoptotic cells, extracellular matrix molecules, and growth factors, and there is mounting evidence to indicate that these proteins are associated with human diseases including susceptibility to pathogens and cancer [17].

ETV6 along with five other genes, BCR, ZMIZ, EML, and Nup21 form chimeric transcripts with ABL1. There must be a joining of the 3’ sequence of ABL1 with the 5’ end of the partner genes, and most of these genes are associated with a wide spectrum hematologic malignancies. Despite this heterogeneity, there is likely a common pluripotent stem cell that gives rise to similar transduction pathways and transforming activity [1]. Due to diagnostic, prognostic, and treatment-related implications, these cases further underscore the use of FISH along with routine chromosome analysis to properly characterize rare, albeit clinically significant fusion genes.

Eosinophilia is a recurrent morphologic finding in ETV6-ABL1-positive myeloid malignancies [4]. Of the five known AML ETV6-ABL1 positive cases, three out of the five were reported to have an increased abnormal eosinophil count, consistent with the findings in the present case. Another finding common to most of the patients was leukocytosis and out of those three cases, two had both leukocytosis and eosinophilia [3, 4, 13, 14]. Each patient was treated with chemotherapy including, cytosine-arabinoside, idarubicin, etoposide, mitoxantrone, and cytarabine. The two patients treated with imatinib responded transiently with resolved fluid retention, eosinophilia, and leukocytosis; however, full remission was not achieved [14]. There was only one patient who achieved full cytogenetic and hematological remission 20 months after undergoing allogeneic hematopoietic stem cell transplantation, which suggests its effectiveness in the treatment of ETV6-ABL1-positive AML patients with eosinophilia and leukocytosis [4].

Although there is limited information about the pathogenesis of myeloid neoplasms positive for ETV6-ABL1, chronic myeloid leukemia (CML) positive for BCR-ABL1 has been well studied and the molecular mechanisms of leukemogenesis and courses of clinical management are established. Tyrosine kinase inhibitors (TKI) such as imatinib are effective agents for inhibiting the constitutively activated BCR-ABL1 tyrosine kinase in CML [18]. Similarly, ETV6-ABL1 is also known to constitutively activate the ABL1 tyrosine kinase, leading to cell cycle deregulation and leukemogenesis [19]. Due to the similar molecular pathogenesis of BCR-ABL1 and ETV6-ABL1 driven leukemogenesis, TKIs have also been considered in the treatment of patients bearing the ETV6-ABL1 fusion. Six out of eleven CML patients positive for ETV6-ABL1 reported in the literature were treated with imatinib: three patients showed a transient favorable response followed by relapse, one patient showed significantly decreased levels of leukemic clones, and two patients treated with 400 mg/day during the chronic phase achieved complete remission [1, 6, 18, 20–22]. Of the three that relapsed, Gancheva et al. reported a case in which the patient was administered an additional TKI, nilotinib, and the patient was able to sustain a positive response following the relapse [1]. Perna et al. did further analysis on another patient who achieved complete remission post-treatment which showed that the ETV6-ABL1 transcript became undetectable, the white blood count normalized, and expression of C-MYC, ID1, BCL-XL, and NUP-98 had decreased significantly [22]. All in all, the molecular targets of ETV6-ABL1 and BCR-ABL1 have significant overlaps that warrant further investigation to elucidate the effectiveness of TKIs on ETV6-ABL1 positive hematologic malignancies.

All in all, this is the sixth reported case of AML bearing the ETV6-ABL1 fusion gene and provides additional insight into the pathogenesis of this subset of malignancies. It is particularly important to utilize complimentary cytogenetic methodologies - namely conventional cytogenetics and FISH - in order to elucidate cryptic abnormalities, which occur more frequently in this context, and to properly characterize karyotypic changes. Additionally, screening using RT-PCR as well as other methodologies has proven useful when cytogenetic analysis is unavailable or yields negative results and in the context of broad molecular screening to identify previously unreported cases. Finally, the consideration of tyrosine kinase inhibitors, particularly second-generation ones, in the treatment of ETV6-ABL1-positive hematological malignancies has shown varying responses, and further investigation of its utility and clinical efficacy is warranted.