KMT2A gene aberrations are more frequent in infants less than one year (yr) of age, accounting for about 70% of acute lymphoblastic leukemia (ALL) patients, and about 30% are diagnosed with acute myeloid leukemia (AML). The most common abnormality found in these patients is the translocation t(11;19)(q23;p13.3), corresponding to 22% of KMT2A rearranged (KMT2A-r) positive cases. Related literature data shows that infants <1 yr with t(11;19)(q23;p13.3) present with up to 50% additional chromosomal changes. In this context, Meyer and coworkers (2007) showed that 25% of the patients with t(11;19)(q23;p13.3) and the KMT2A-MLLT1 fusion transcript presented three-way or more complex fusions, associated with a worse prognosis. In such cases, KMT2A-r must be assessed by more sensible techniques combining high-resolution cytogenetics and molecular approaches, providing a precise characterization for the correct choices in risk-adapted therapy. In this work, we show a cryptic three-way translocation t(9;11;19) harboring a KMT2A-MLLT1 and the novel SEC16A-KMT2A fusion, associated with SEC16A and KMT2A aberrant expression levels, in an infant with B-ALL presenting a poor prognosis.

At diagnosis, a bone marrow sample from a 2-month-old boy was referred to the Laboratory of Cytogenetics – INCA. Immunophenotyping showed 94% of blast cells presenting lymphoid morphology and the diagnosis of pro-B lymphoblastic, being treated under a high-risk BFM INTERFANT 06 protocol. The patient could not experience any treatment response and died due to an early relapse. G-banding and FISH were performed on bone marrow and peripheral blood samples under standard protocols. LDI-PCR assays were used to identify the KMT2A partner genes and their corresponding breakpoints. RT-qPCR analyses were performed to verify the levels of transcript expression of genes involved in the rearrangements.

G-banding showed the karyotype: 47,XY,+X,t(9;11)(q34;q23). The FISH analysis revealed a KMT2A-r. Molecular cytogenetic analyses revealed a cryptic three-way translocation. The LDI-PCR sequencing revealed a KMT2A-MLLT1 and the novel out-of-frame SEC16A-KMT2A . RT-qPCR showed standard expression levels for MLLT1, whereas revealed a SEC16A reduced expression (75%), and KMT2A overexpression (3-fold).

To relate the abovementioned data to the patient's clinical presentation, we hypothesized relation between the expression profiles of the fusion transcripts and the patient's dismal prognosis. Thus, we compared the relative expression among three different KMT2A regions, donor X patient. The first region, spanning exons 3 and 4, is present in both samples and was attributed the value 1. Spanning exons 11 and 12 (SEC16A-KMT2A fusion region), despite its lower expression in the patient, no significant difference was observed. On the other hand, spanning exons 14 and 15 (downstream transcript of the KMT2A fusion) was significantly lower expressed in the patient (0.4-fold). These results confirm that increased KMT2A expression in the patient depends on the wild-type allele and that the KMT2A transcripts with fusions are probably expressed.

Our case illustrates the importance of molecular tests in selecting cases for further investigations of posttranscriptional regulation mechanisms, which could open perspectives regarding novel therapeutic approaches for poor prognosis childhood leukemias.