Involvement of the AML1 gene in the t(3;21) in therapy-related leukemia and in chronic myeloid leukemia in blast crisis

A nonrandom translocation between chromosomes 3 and 21, t(3;21)(q26.2;q22) has been detected in patients with a myelodysplastic syndrome or acute myeloid leukemia after treatment (t-MDS/t-AML) for a primary malignant disease and in chronic myelogenous leukemia in blast crisis (CML-BC). In these patients, the breakpoint on chromosome 21 is at band 21q22. This band is also involved in the t(8;21)(q22;q22) detected in 40% of the patients with acute myeloid leukemia subtype M2 (AML-M2) de novo who have an abnormal karyotype. In the t(8;21), the AML1 gene is the site of the breakpoint on chromosome 21. The AML1 gene is transcribed from telomere to centromere, and in the t(8;21) the 5' part of AML1 is fused to the ETO gene on chromosome 8 to produce the chimeric AML1/ETO on the der(8) chromosome. We found that AML1 is also rearranged in two t-AML patients and in one CML-BC patient with the t(3;21), but the breakpoints are approximately 40 to 60 kb downstream to those of AML-M2 patients. This region contains at least one additional exon of AML1, as determined by using an AML1 cDNA as a probe in Southern blot analysis. The t(3;21) breakpoints for the remaining patients could not be determined because, by fluorescence in situ hybridization analysis, the breaks are outside of the region covered by the available probes.     

The 3;21 translocation in myelodysplasia results in a fusion transcript between the AML1 gene and the gene for EAP, a highly conserved protein associated with the Epstein-Barr virus small RNA EBER 1.

In the 8;21 translocation, the AML1 gene, located at chromosome band 21q22, is translocated to chromosome 8 (q22), where it is fused to the ETO gene and transcribed as a chimeric gene. AML1 is the human homolog of the recently cloned mouse gene pebp2 alpha B, homologous to the DNA binding alpha subunit of the polyoma enhancer factor pebp2. AML1 is also involved in a translocation with chromosome 3 that is seen in patients with therapy-related acute myeloid leukemia and myelodysplastic syndrome and in chronic myelogenous leukemia in blast crisis. We have isolated a fusion cDNA clone from a t(3;21) library derived from a patient with therapy-related myelodysplastic syndrome; this clone contains sequences from AML1 and from EAP, which we have now localized to band 3q26. EAP has previously been characterized as a highly expressed small nuclear protein of 128 residues (EBER 1) associated with Epstein-Barr virus small RNA. The fusion clone contains the DNA binding 5' part of AML1 that is fused to ETO in the t(8;21) and, in addition, at least one other exon. The translocation replaces the last nine codons of AML1 with the last 96 codons of EAP. The fusion does not maintain the correct reading frame of EAP and may not lead to a functional chimeric protein.     

Minimal residual disease studies are beneficial in the follow-up of TEL/AML1 patients with B-precursor acute lymphoblastic leukaemia

The t(12;21)(p13;q22) translocation has been identified as the most common chromosomal abnormality in childhood acute lymphoblastic leukaemia (ALL). Initially, several investigators reported an excellent prognosis in paediatric leukaemias with this translocation, but other studies showed a 20% incidence in relapsed ALL. We performed an extensive analysis of 90 ALL patients. In 17 (19%) cases a TEL/AML1 fusion was found. However, this group was not representative as it included a high number of relapsed patients compared with the normal incidence in B-precursor ALL [54 in continuous complete remission (CCR) and 36 relapsed patients] and only a slightly better prognosis for TEL/AML1-positive patients was found (not significant) (four relapses in 17 TEL/AML1-positive patients vs. 32 relapses in 73 TEL/AML1-negative patients). Comparison of known prognostic factors (age, sex, ploidy, white blood cell count and immunophenotype) between relapsed TEL/AML1-positive and TEL/AML1-positive patients in CCR did not reveal differences, except that the white blood cell count was significantly higher in the relapsed group (P = 0.001). Time between diagnosis and relapse was not different for the relapsed TEL/AML1-positive group vs. the relapsed TEL/AML1-negative group. In 11 TEL/AML1-positive patients, the minimal residual disease (MRD) level at the end of induction therapy was quantified in a limiting dilution assay using IGH or TCRD junctional regions as polymerase chain reaction (PCR) targets. In all four relapsed patients, the level of MRD at the end of induction therapy was high (range 0.24-1.2%), whereas in all seven CCR patients, the MRD level was extremely low (0.02 to < 0.001%). In agreement with previous studies in which MRD levels at the end of induction therapy were found to be the strongest risk factor independent of other risk factors, in the present study we show that the MRD level remains a risk factor independent of the presence of a TEL/AML1 fusion gene.