The 'promiscuous' E2A gene, at 19p13.3, is fused with two different molecular partners, PBX1 and HLF, following two chromosome translocations recurrent in childhood pre-B ALL. We have identified a novel gene, FB1, by virtue of its fusion with E2A and by a combination of molecular techniques. FB1 was localized on 19q13.4, suggesting that the novel chimera originated by a cryptic rearrangement of chromosome 19. Two FB1 transcripts, of 1.2 kb and 1.1 kb, are differentially expressed at low level in a variety of human tissues, including hemopoietic cell lines from different lineages. Accordingly, FB1 cDNA displays high homology with a number of cDNA clones from different human tissues. High homology was found also with cDNA clones from mouse and rat, suggesting that the sequence might be conserved at least among mammals. The function of the putative FB1 protein, however, is currently unknown as database sequence comparisons have failed to reveal strong homology with known proteins. The E2A/FB1 fusion appears to be a recurrent feature of pre-B ALLs, suggesting that it might have a role in the development and/or progression of leukemogenesis.

The prognostically important 1;19 chromosomal translocation can alter the E2A gene on chromosome 19p13 in childhood B-cell precursor acute lymphoblastic leukemia (ALL), leading to formation of a fusion gene (E2A-PBX1) that encodes a hybrid transcription factor with oncogenic potential. It is not known whether this molecular alteration is a uniform consequence of the t(1;19) or is restricted to translocation events within specific immunologic subtypes of the disease. Therefore, we studied leukemic cells from 25 cases of B-cell precursor ALL, with or without evidence of cytoplasmic Ig mu heavy chains (cIg); 17 cases had the t(1;19) by cytogenetic analysis. Leukemic cell DNA samples were analyzed by Southern blotting to detect alterations within the E2A genomic locus; a polymerase chain reaction assay was used to identify expression of chimeric E2A-pbx1 transcripts in leukemic cell RNA; and immunoblotting with anti-Pbx1 antibodies was used to detect hybrid E2A- Pbx1 proteins. Of 11 cases of cIg+ ALL with the t(1;19), 10 had E2A- pbx1 chimeric transcripts with identical junctions and a characteristic set of E2A-Pbx1 hybrid proteins. Each of these cases had E2A gene rearrangements, including the one in which fusion transcripts were not detected. By contrast, none of the six cases of t(1;19)-positive, cIg- ALL had evidence of rearranged E2A genomic restriction fragments, detectable E2A-pbx1 chimeric transcripts, or hybrid E2A-Pbx1 proteins. Typical chimeric E2A-pbx1 transcripts and proteins were detected in one of eight cIg+ leukemias in which the t(1;19) was not identified by cytogenetic analysis, emphasizing the increased sensitivity of molecular analysis for detection of this abnormality. We conclude that the molecular breakpoints in cases of cIg- B-cell precursor ALL with the t(1;19) differ from those in cIg+ cases with this translocation. Leukemias that express hybrid oncoproteins such as E2A-Pbx1 or Bcr-Abl have had a poor prognosis in most studies. Thus, molecular techniques to detect fusion genes and their aberrant products should allow more timely and appropriate treatment of these aggressive subtypes of the disease.

Several partner genes on different chromosomes have been reported to be fused with the ETV6 gene (located in chromosome band 12p13), with different breakpoints and different frequencies, in various hematologic malignancies, particularly acute myeloid and lymphoid leukemias and myelodysplastic syndromes. By using FISH and molecular analyses, we have analyzed five different pediatric and adult patients carrying cytogenetic abnormalities involving 12p13. Our findings demonstrate that ETV6 was rearranged in all the cases analyzed. In particular, ETV6 was disrupted by translocations with chromosomal bands 7q22, 7q36, 9q11, and 13q12, not previously described as partners of ETV6 in translocations, thus extending its promiscuity in rearranging with different partner genes.

We identified seven new recurring translocations among 483 cases of acute lymphoblastic leukemia (ALL) with adequate chromosome banding studies. Four were apparently balanced [t(1;3)(p34;p21), t(7;9)(p15;p23- p24), t(12;13)(p13;q14), t(17;19)(q22;p13)], while three were unbalanced with the formation of a dicentric chromosome [dic(7;9)(p13;p11), dic(7;12)(p11;p12), and dic(12;17)(p11;p11-p12)]. One translocation was observed in five cases, two in four cases, and the remaining four in two cases each. The modal chromosome numbers in these 21 cases were 45 (n = 11), 46 (n = 8), and 47 (n = 2). Eight of the 11 cases with a dicentric chromosome had a modal number of 45. Only a single translocation was found in 14 cases (67%), representing the sole structural abnormality in six cases. In three of the seven translocation subgroups, the blast cells were consistently of B lineage (pre-B, early pre-B, or both); in all others, they represented both the B and T lineages. The small size of these subgroups prevented definitive clinical correlations, although it may be important that two of the four cases with a t(17;19) and an early pre-B-cell immunophenotype had disseminated intravascular coagulation, an event usually observed in acute promyelocytic leukemia or T-cell ALL. These findings add substantially to the existing list of nonrandom chromosomal translocations in childhood ALL and may help to explain the genetic alterations leading to the loss of normal growth control mechanisms in this disease.

The t(1;19) is the most frequent recurring chromosomal translocation in childhood acute lymphoblastic leukaemia (ALL). In most cases typical chimaeric E21-PBX1 transcripts are expressed as a consequence of this rearrangement, allowing the molecular detection of the t(1;19) at the RNA level. This translocations has been associated with a poor clinical outcome, although intensified chemotherapy has been reported to nullify its adverse prognostic impact. We therefore used reverse transcriptase/polymerase chain reaction (RT-PCR) to detect residual leukaemic cells at successive times during treatment and to monitor the response to chemotherapy in six t(1;19)-positive ALL pediatric patients. Five of these patients rapidly achieved molecular remission and no evidence of minimal residual disease (MRD) was found in the remission bone marrows beyond the third month of treatment. One patient still displayed residual leukaemic cells at the end of therapy, although she has been in continuous complete clinical remission (CCR) for 84 months. However, this patient is peculiar in our series in that two different types of chimaeric E2A-PBX1 transcripts were expressed in her leukaemic cells, only one being detectable in remission.