ETV6-RUNX1 Rearrangement in Tunisian Pediatric B-Lineage Acute Lymphoblastic Leukemia

In this study, Forty-one out of fifty-seven Tunisian children with B-lineage acute lymphoblastic leukemia (B-ALL), and without cytogenetically detectable recurrent abnormalities at the time of the diagnosis, were evaluated by fluorescence in situ hybridization (FISH) for the t(12;21). This translocation leads ETV6-RUNX1 (previously TEL-AML1) fusion gene. 16 patients (28%) had ETV6-RUNX1 rearrangement. In addition to this rearrangement, two cases showed a loss of the normal ETV6 allele, and three others showed an extra signal of the RUNX1 gene. Seven patients without ETV6-RUNX1 rearrangement showed extra signals of the RUNX1 gene. One out of the 7 patients was also associated with a t(3;12) identified by FISH. This is the first Tunisian study in which we report the incidence of t(12;21) among childhood B-lineage ALL and in which we have found multiple copies of RUNX1. Finally, our findings confirm that additional or secondary genetic changes are commonly encountered in pediatric B-lineage ALL with ETV6-RUNX1 gene fusion which is envisaged to play a pivotal role in disease progression.


Introduction
The t(12;21)(p13;q22) translocation is the most common genetic alteration in childhood B-cell acute lymphoblastic leukemia occurring in approximately 25% of cases and is associated with a favorable outcome [1,2].
The molecular consequence of this translocation is the fusion of two known genes: ETV6 mapped to 12p13 and RUNX1 located at 21q22 [1,3]. It results in the formation of a hybrid protein involving the helix-loop-helix domain of ETV6 and the entire RUNX1 gene that appears to interfere with transactivation by the normal RUNX1 in a transdominant manner [3,4].
Because it is a cryptic translocation which usually escapes diagnosis on conventional cytogenetic (CC) study due to the similarity of the changed bands, molecular cytogenetic tools such as fluorescence in situ hybridization (FISH) are needed to determine the incidence of the ETV6-RUNX1 fusion in the Tunisian B-lineage ALL pediatric cases that had normal karyotypes or random chromosome aberrations at diagnosis and to compare our findings to those previously described in literature.

Conventional Cytogenetics.
Chromosomal preparations with RHG-banding were performed according to a previously described protocol [5].  Chromosomes were identified and arranged according to the international system for human cytogenetic nomenclature (ISCN 2005) [6], and the karyotype profile was determined by the analysis of at least 20 metaphases. (FISH). The presence of the ETV6-RUNX1 fusion gene was assessed using two YAC clones (the YAC936E2 and the YAC821S11) selected from their location in the 12p13 (ETV6) and the 21q22 (RUNX1) regions, respectively, in the "Centre d'Etude du Polymorphisme Humain Database" (http://www.cephb.fr/).
After drying, the ETV6-RUNX1 dual-color FISH probe, denatured as preliminary at 75 • C for 5-10 minutes, was applied into a slide, covered with a 24 × 24 mm coverslip and sealed with a rubber cement. Slides were then placed in a humid light-proof container at 37 • C for overnight hybridization.
After hybridization, the coverslip was removed and slides were washed in 0.4 standard saline citrate (SSC) for 5 minutes at 75 • C, followed by a second wash in 2XSSC-0.1% NP40 for 2 minutes.
After drying, the slides were counterstained with 4' ,6diamidino-2-phenylindole and examined with a fluorescent microscope equipped with appropriate filters and CytoVision FISH system image capture software (Zeiss Axioskop 2 plus) and at least 50 metaphase cells, and 100 interphase nuclei were analyzed for ETV6-RUNX1 translocation.
In a normal case, the hybridization with the ETV6 (red) and the RUNX1 (green) probes showed two red and two green signal patterns (Figure 1(a)). However, in a case with the t(12;21), the hybridization with these probes showed one or two fusion signals (red/green or yellow), one red and one green signal patterns corresponding to the normal copies of the ETV6 and the RUNX1 genes, respectively, (Figure 1).
The selected 41 cases included 25 males and 16 females with age ranging from 1.2 to 15 years (mean 7.4 years).
Additional chromosomal aberrations accompanying the t(12;21) were found in five patients (31.2%). Deletion of nontranslocated ETV6 allele was detected in two patients (cases 1 and 2) (Figure 1(c)), and three patients without +21 showed extra signal of RUNX1 (cases 4, 14, and 21) (Figure 1(d)). Those include a case in which the clone with the extra signal of RUNX1 was distinct from the one harboring ETV6-RUNX1 fusion gene (case 21, Table 1).
Extra copies of RUNX1 gene were also detected in 7 cases without t(12;21); three of which (cases 8, 17, and 23) do not have +21 on the karyotype. In the four cases remaining (cases 5, 6, 7, and 16) the number of RUNX1 signals detected by FISH was in agreement with the number of chromosome 21 or the presence of marker chromosomes (Figure 1(e)). Of the cases, one (case 5) was associated with t(3;12) detected by metaphase FISH (Figure 1(f)).

Discussion
In the present study, we used FISH for the detection of ETV6-RUNX1 fusion gene in 41 children with B-lineage ALL that had normal karyotypes or random chromosome aberrations at diagnosis. To our knowledge, this is the first report of the frequency of the t(12;21) from Tunisia and the second from an Arab country after the Egyptian study published by Mikhail et al. in 2002 [7].
We confirm so the efficiency of FISH technique in the detection of such cryptic chromosomal rearrangement, usually missed by CC [10,11].
However, in B-lineage ALL patients of Germany [11], Italy [11], and France [15], the frequency of this translocation is reported to be about 25%. A frequency of 31% is found among American patients [16], and we recently report in Israel [17] and in Brazil [8] an incidence of 40%.
This difference could be explained by the fact that due to the expected peak incidence at 3-5 years reported in the majority of the published series, our cases could have been biased for old age (mean age 7.4 years), which then corresponds to lower t(12;21) incidence.
The ETV6-RUNX1 fusion gene on the der(21) was detected in all patients with t(12;21), whereas the reciprocal ETV6-RUNX1 fusion gene on the der (12) was observed in only two cases.
While this t(12;21) may initiate the leukemic process, critical secondary genetic events found in association with ETV6-RUNX1 fusion gene are currently believed to be pivotal for leukemogenesis and are previously reported in literature [4,[18][19][20].
In our study, additional chromosomal aberrations were found in five patients among ETV6-RUNX1 fusion positive cases (31.2%).
This low detection of del (12) in our study could be explained by the fact that the ETV6 FISH probe covered almost the entire of the ETV6 gene so that submicroscopic deletion could be missed.
According to the Knudson' hypothesis [9], ETV6 could act as a tumor suppressor gene, with t(12;21) disrupting one copy of ETV6 and the deletion being the second inactivating event [17,21].
In the three other patients, we have found extra copies of the RUNX1 gene in which the clone with the extra signal of RUNX1 was distinct from the one harboring ETV6-RUNX1 fusion gene.
Interestingly, extra copies of RUNX1 gene were also observed in 7 patients without ETV6-RUNX1 fusion. Of them, three do not have +21 on their karyotype which is likely consistent with extra 21 perhaps in a small clone not observed cytogenetically or derive from non dividing clonal hyperdiploid cell, while normal metaphases derived from normal dividing cells.
In the four remaining cases, the number of RUNX1 signals was in keeping with results on CC including a case (case 5) in which we have identified a t(3;12) by FISH. This translocation has previously been reported in a case with Blineage ALL but in association with t(12;21) by Yehuda-Gafni et al. [17] and Kobayashi et al. [19].
Extracopies of the RUNX1 gene were found in 6 patients (having or not ETV6-RUNX1 fusion gene and without +21 in the karyotype) among 57 patients with B-lineage ALL (10.5%). This incidence was higher than that previously reported by Dal Cin et al. [24], Niini et al. [25] and more recently by Busson Le-Coniat et al. [26] who reported only 2 or 3 cases.
Although the real significance of these findings has not been clarified yet, the hypothesis of the gene-dosage effect involving RUNX1 seems to be very probable [27,28].
In summary, our findings demonstrate that ETV6-RUNX1 fusion gene is a common genetic abnormality detected by FISH in 28% of Tunisian children with Blineage ALL, and we confirm that secondary genetic events are commonly encountered in these patients. Interestingly, extra copies of RUNX1 are frequently found in our series than those previously reported in other populations and so warrant further investigation to elucidate the mechanisms underlying the role of the RUNX1 in leukemogenesis. Moreover, due to the unknown prognostic significance of this abnormality, further studies should be conducted in consecutive children with ALL to correlate RUNX1 overexpression with the patients' followup.