A case of B-cell acute lymphoblastic leukemia in a child with Down syndrome bearing a t(2;12)(p12;p13) involving ETV6 and biallelic IGH@ rearrangements

Background Rearrangements involving ETV6 (12p13) are among the most common structural abnormalities in pediatric B-cell acute lymphoblastic leukemia (B-ALL) and involve numerous partner genes. Additionally, the t(8;14)(q11.2;q32), which can result in the placement of CEBPD (8q11.2) near the regulatory regions of IGH@ (14q32) and consequent overexpression of CEPBD, occurs at a higher frequency in individuals with Down syndrome-associated ALL (DS-ALL) compared to both the general and pediatric population. The coexistence of cytogenetically detectable ETV6 abnormalities and t(8;14)(q11.2;q32) is a rare occurrence in B-ALL and has only been reported in a single case in the literature. Findings Herein, we present a case of B-ALL in a 9-year old male with Down syndrome in which conventional cytogenetic analysis revealed two reciprocal translocations: a t(8;14)(q11.2;q32) and a t(2;12)(p12;p13). Interphase and metaphase fluorescence in situ hybridization (FISH) analysis using break apart probes confirmed the involvement of IGH@ and ETV6 in these translocations, respectively. Additionally, interphase FISH revealed a clonal subpopulation bearing biallelic IGH@ rearrangements not observed by conventional cytogenetic analysis. Conclusions To the best of our knowledge, this is the first reported case of B-ALL bearing an ETV6 translocation with a partner gene on the short arm of chromosome 2 confirmed by FISH. Additionally, it is the second reported case of t(8;14)(q11.2;q32)-ALL bearing a concomitant, cytogenetically detectable abnormality involving ETV6. This case provides insight into a novel translocation involving ETV6 as well as potentially unique and understudied mechanisms of clonal evolution in pediatric B-ALL.

Rearrangements involving ETV6 (12p13) are among the most common structural abnormalities in pediatric B-ALL. The t(12;21)(p13;q22), the most common of these translocations, results in the production of a chimeric transcription factor bearing the DNA-binding domain of RUNX1 (21q22) and the transactivation domain of ETV6 (12p13), resulting in aberrant activation of genes regulated by RUNX1. Although the t(12;21) [ETV6/RUNX1] is the most common of these rearrangements, other translocations involving ETV6 with greater than 20 partner genes have been observed, including protein tyrosine kinases and transcription factors, many of which can act by distinct mechanisms to promote leukemogenesis [3]. Additionally, other anomalies involving ETV6 have been observed in various hematological malignancies ranging from deletions, to point mutations, to alterations at the epigenetic level, to amplifications [3][4][5]. Although abnormalities involving ETV6 are a relatively common finding in B-ALL, the precise leukemogenic role of the gene in the context of some of the aforementioned aberrations remains understudied.

Case presentation
A 9-year old male with Down syndrome presented with persistent fever and fatigue. Complete blood count revealed pancytopenia (WBC 3.01×10 3 /μL, RBC 2.87×10 6 / μL, platelet count 43×10 3 /μL) with a differential of 32 % lymphoblasts, 52 % lymphocytes, 7 % neutrophils, 4 % monocytes, 1 % metamyelocyte, and 1 % myelocyte. Flow cytometry on peripheral blood revealed excess abnormal blasts comprising 22 % of total cells, and expressing CD34, CD10, CD19, CD22, and HLA-DR. A bone marrow biopsy showed hypercellular marrow (~90 %) and 95 % replacement by sheets of lymphoblasts. These findings are consistent with a diagnosis of B-lymphoblastic leukemia, and thus, a diagnosis of B-ALL was rendered. Induction chemotherapy was immediately started with Vincristine and cytarabine. On day 29 post induction chemotherapy, a bone marrow biopsy showed variably lower cellular marrow with approximate overall cellularity of 80 %. A followup bone marrow biopsy showed minimal residual disease, displaying a favorable response to therapy.

Conventional cytogenetics
Chromosome analysis was performed on 30 metaphase spreads from bone marrow and peripheral blood using standard cytogenetic techniques. Karyotypes were prepared using Applied Imaging CytoVision software (Applied Imaging, Genetix, Santa Clara, CA) and described according to the International System for Human Cytogenetic Nomenclature (ISCN) 2013 [6].

Fluorescence in situ hybridization (FISH)
FISH was performed on interphase nuclei and/or previously G-banded metaphase spreads using the following probes acquired from Abbott Molecular (Abbott Molecular, Des Plaines, Illinois 60018):

Fluorescence in situ hybridization (FISH)
FISH analysis on interphase nuclei using the Vysis LSI ETV6/RUNX1 ES Dual Color Translocation Probe Set revealed 3 signals corresponding to ETV6 in 50.7 % (152/300) of the nuclei examined, suggesting a potential ETV6 (12p13) gene rearrangement (Fig. 2). All 300 nuclei examined exhibited 3 signals corresponding to RUNX1 (21q22), consistent with the constitutional genetic makeup of this patient (Down syndrome). FISH on previously G-banded metaphase spreads using the Vysis LSI ETV6 Dual Color, Break Apart Probe Kit revealed a fusion signal (F) corresponding to an intact ETV6 gene on the normal chromosome 12 and a split signal (1R1G) corresponding to a rearranged ETV6 gene, with a red signal (5′/telomeric region of ETV6) on the derivative  FISH analysis on interphase nuclei using the Vysis LSI IGH Dual Color, Break Apart Rearrangement Probe revealed two abnormal clonal populations (Fig. 4). The first clone showed one fusion signal and one split signal (1F1R1G) and was observed in 269/300 (89.7 %) of the nuclei examined. The second clone showed two split signals (2R2G) and was observed in 20/300 (6.7 %) of the nuclei examined. The former suggests a rearrangement involving one IGH@ gene with the other remaining intact (monoallelic), and the latter suggests rearrangements of both IGH@ genes (biallelic). FISH on previously G-banded metaphase spreads also revealed the former signal pattern (1F1R1G) consisting of a fusion signal and a split signal (1F1R1G), with the fusion signal corresponding to an intact IGH@ gene on the normal chromosome 14, the red signal (3′/ telomeric region of the IGH@ gene) on the derivative chromosome 8, and the green signal (5′/centromeric region of the IGH@ gene) on the derivative chromosome 14, ultimately confirming the t(8;14) observed by conventional cytogenetic analysis (Fig. 5). No metaphase spreads bearing a second 14q32 rearrangement were observed by conventional cytogenetic analysis, and we were thus not able to detect or localize the latter signal pattern (2R2G; two split signals) by metaphase FISH.
Analysis using other probes did not reveal any additional abnormal signal patterns. This constellation of results was described as follows: nuc

Conclusion
The cytogenetic findings in this case highlight a unique combination of rearrangements that has not been previously described in B-ALL. The t(8;14)(q11.2;q32) is a recurrent translocation in B-ALL that generally causes exchange of the regulatory regions of the IGH@ gene and the CEBPD gene, placing CEBPD in close proximity to the regulatory regions of IGH@ and resulting in its overexpression [7]. Although it is relatively rare in the general population, this translocation is more frequent in Down syndrome-associated ALL (DS-ALL) and is associated with a B-cell precursor immunophenotype and an intermediate prognosis [2,8,9]. It has also been shown to be associated with a number of secondary cytogenetic abnormalities, including der(14)t(8;14)(q11.2;q32), t(9;22)(q11.2;q34), +21, +X, and +14, although patients with Down syndrome and t(8;14)(q11.2;q32)-positive ALL generally do not bear a concomitant Philadelphia chromosome [2]. The t(2;12)(p12; p13) is a recurrent, albeit rare, translocation in cyclin D1-negative mantle cell lymphoma (MCL). In these cases, it involves the IGK (2p12) and CCND2 (12p13) genes, and results in overexpression of CCND2 due to the placement of the CCND2 gene near the regulatory regions of the IGK gene [10]. However, in our case, FISH analysis using both a break apart probe and a dual color translocation probe specific to the ETV6 gene, which is also located at the 12p13 band, confirmed that the t(2;12)(p12;p13) we observed involves the ETV6 gene, and not CCND2. Additionally, since both ETV6 and CCND2 localize to 12p13, we confirmed that the ETV6 break apart probe (Vysis LSI ETV6 Dual Color, Break Apart Probe Kit) we utilized in our analysis bears no overlap with the CCND2 gene, minimizing the possibility of a false-positive result. The break apart probe we utilized spans chr12: 11321260-12578058, whereas the CCND2 gene spans chr12: 4382902-4414522. We queried the Mitelman Database of Chromosome Aberrations and Gene Fusions in Cancer for cases of B-ALL bearing a t(2;12)(p11-p13; p12-p13) and were only able to identify seven cases in addition to our case, indicating that it is an extremely rare cytogenetic abnormality in B-ALL ( Table 1). None of these cases included corresponding molecular analysis of ETV6, so the involvement of the gene in the reported cases is not known. Thus, to the best of our knowledge, this is the first reported case of B-ALL bearing a t(2;12)(p12;p13) with ETV6 involvement confirmed by molecular analysis. Of note, each of the previously reported cases and our case occurred in pediatric patients, and five out of eight of the cases carried chromosome 21 abnormalities, including constitutional and/or acquired +21 and i(21q).
ETV6 is able to promote leukemogenesis via a variety of mechanisms. Most commonly, ETV6 is involved in translocations with partner genes that encode protein tyrosine kinases or transcription factors, leading to deregulation of signaling pathways essential to hematopoiesis via generation of fusion genes or aberrant activation of protooncogenes [3]. Additionally, most evidence supports the notion that ETV6 functions as a tumor suppressor gene. For example, in cases where one allele of ETV6 is involved in a translocation, there is often a concomitant deletion of the other, non-rearranged allele [3]. Additionally, decreased or absent ETV6 expression has been observed in cases that don't bear a deletion of ETV6. Finally, point mutations leading to loss of function of ETV6 have also been observed [3]. These lines of evidence suggest that the t(2;12)(p12; p13) in our case is likely a translocation involving ETV6 and an unknown partner gene on the short arm of chromosome 2 leading to aberrant activation of a protein tyrosine kinase pathway or a proto-oncogene encoded by that unknown partner gene. Alternatively, this rearrangement could result in loss of function of ETV6 without affecting or involving a partner gene, which has also been shown to contribute to leukemogenesis [3].
However, two reported casesone case of myelodysplastic syndrome and one case of B-ALLshowed amplification of the ETV6 gene via the generation of homogeneously staining regions (hsr) on the short arm of chromosome 12 [4,5]. In both of these cases, further molecular analysis confirmed that the homogeneously staining regions consisted primarily of ETV6 gene material, confirming the amplification, and immunohistochemical analysis confirmed overexpression of ETV6 compared to case-matched controls [4,5]. Furthermore, no additional mutations involving ETV6 that are known or predicted to result in its loss of function were observed in these cases. These two cases provide evidence against the hypothesis that ETV6 functions exclusively as a tumor suppressor gene and suggests that in certain contexts, it can function as an oncogene via its overexpression, a rare phenomenon that has been observed in other genes such as the Wilms tumor 1 (WT1) gene [11]. Because the t(2;12)(p12; p13) observed in MCL involves overexpression of CCND2 due to placement of the gene near the regulatory regions of IGK, it is possible that in our case, a similar mechanism could result in overexpression of ETV6 in the context of this translocation due to the placement of ETV6 near the regulatory regions of IGK. Additionally, Lu et al. reported the first case of B-ALL bearing a t(12;14)(p13;q32) involving ETV6 and IGH@ confirmed by metaphase FISH using probes specific to both genes, showing that translocations involving ETV6 and immunoglobulin genes can occur in B-cell neoplasias [12]. However, such immunoglobulin translocations can also result in mutation of the partner gene in addition to aberrant expression, which can affect the wild-type function of a tumor suppressor gene [13]. Unfortunately, due to limited sample material, we were not able to conduct further molecular or functional analysis of this case to determine the precise nature of the t(2;12) and the role of ETV6 in the evolution of this malignancy.
The variation of abnormalities of the IGH@ and ETV6 genes and the variation in the frequencies of these abnormalities as detected by interphase FISH provides evidence for an underlying cytogenetic evolution in the development of this malignancy. First of all, clones bearing a monoallelic IGH@ rearrangement, evidenced by a 1F1R1G signal pattern by interphase FISH using an IGH@ break apart probe, were observed in 89.7 % of interphase nuclei analyzed. Conversely, only 6.7 % of cells showed evidence of biallelic IGH@ rearrangements (2R2G) using the same probe and 50.7 % of interphase nuclei showed evidence of an ETV6 rearrangement when analyzed with a dual color, dual fusion translocation probe, suggesting that a monoallelic IGH@ rearrangement, caused by t(8;14)(q11.2;q32), was the primary cytogenetic abnormality in this malignancy. Because we were not able to detect biallelic 14q32 rearrangements by conventional cytogenetic analysis but were able to observe both a monoallelic t(8;14)(q11.2;q32) and t(2;12) (p12;p13) in the same metaphase spread, the evolution of this malignancy likely happened via one of two mechanisms: (1) acquisition of a primary t(8;14), followed by t(2;12) in the same clone, followed by an additional IGH@ rearrangement in the same clone or (2) acquisition of a primary t (8;14), followed by t(2;12) in one clone and an additional IGH@ rearrangement in a different clone (Fig. 6). The former mechanism, depicted in Fig. 6a, would result in a linear cytogenetic evolution such that a clonal population would exist that bears all three abnormalities observed in this case: a t(8;14), a t(2;12), and an additional IGH@ rearrangement. Conversely, the latter mechanism, depicted in Fig. 6b, would result in a divergent cytogenetic evolution  such that one clonal population would bear a t(8;14), a t(2;12), and no additional IGH@ rearrangement and the other would bear a t (8;14) and an additional IGH@ rearrangement, but not a t(2;12). It is important to note that because we did not observe biallelic 14q32 rearrangements in any metaphase spreads by conventional cytogenetic analysis or metaphase FISH, the additional IGH@ rearrangement could be due to an additional t(8;14)(q11.2; q32) or an IGH@ rearrangement involving a different gene or locus. Due to the relatively low frequency of the clone bearing biallelic IGH@ rearrangements, it is more likely that this malignancy evolved via the former, linear mechanism, although both possibilities must be considered. Among all reported cases of t(8;14)(q11.2; q32)-ALL, only one case, reported by Harrison et al., was found to bear a concomitant cytogenetic abnormality involving the 12p13 locus, described as follows: 47, XX, t(8;14)(q11; q32), del(12)(p12p13),+21c. Further FISH studies confirmed both a monoallelic IGH@ rearrangement and a monoallelic ETV6 deletion [14]. It is evident that the involvement of cytogenetically detectable abnormalities involving ETV6 in the evolution of t(8;14)(q11.2; q32)-ALL is an extremely rare occurrence that presents a number of questions regarding molecular mechanisms and clinical implications that remain understudied. We emphasize the importance of utilizing both conventional cytogenetic and molecular genetic tools to elucidate relevant abnormalities and molecular mechanisms in such cases, ultimately to determine the clinical implications of rare cytogenetic abnormalities in pediatric ALL such as those presented in this case. In summary, this case provides insight into a novel translocation involving ETV6 as well as potentially unique and understudied mechanisms of clonal evolution in pediatric B-ALL that warrant further investigation.