Background

Over 70 KMT2A partner genes have been reported so far in the literature. Translocation of the KMT2A gene, better known as the MLL gene, is present in 7–10% of B acute lymphoblastic leukemias, AFF1 (or AF4) being the most frequent partner gene. The prognosis of B acute lymphoblastic leukemias (B ALL) with KMT2A gene translocations are usually poor, especially with KMT2A-AFF1 fusions [1, 2]. One of its rare partner genes, mastermind like 2 (MAML2) gene has been reported in four cases of myeloid neoplasms after chemotherapies: two cases of acute myeloid leukemia (AML) and two cases of myelodysplasic syndromes (MDS) [3]. Secondary acute lymphoid leukemia is a well-documented occurrence following topoisomerase II inhibitors and cyclophosphamide chemotherapy and/or radiotherapy. The premise that radiotherapy as well as chemotherapy are inducing t(4;11)(q21;q23) was raised before [4]. Indeed two cases of secondary T-cell acute lymphoblastic leukemia with KMT2A-MAML2 transcripts have been described in two adolescent leukemia patients [5].

At the molecular level, much progress has been made in the identification of fusion transcripts and finding new alternatives to fluorescent in situ hybridization (FISH). In fact, the identification of fusion transcripts is possible with the Next-Generation Sequencing (NGS) technologies, such as RNA-Sequencing [6]. At the moment, this method seems to be the best for identifying new gene fusion transcripts, and is steadily improving [7]. In this case, we highlight one of the limitations of this powerful tool.

Case presentation

Patient’s history and clinical features

A 47-year-old female, with morbid obesity and hypertrophic cardiopathy, presented to hospital in cardiopulmonary arrest most likely due to a cardiac rhythm disturbance. During hospitalization in the cardiac section, laboratory analysis revealed anemia (Hb 105 g/L), thrombocytopenia (59 × 109 platelets/L) and a total leukocyte count of 6.2 × 109/L with 1% blast cells. Bone marrow aspiration showed hyper cellular marrow with agranular leukemic cells (35%). The examination of the bone marrow morphology revealed dysplasia on the three cell lineages. Flow cytometric immunophenotyping of bone marrow identified blast cells that were positive for cCD79a, CD22, CD19 and CD45, and negative for CD10. There was a history of high-grade thoracic sarcoma in the previous year, treated with surgical resection, three cycles of chemotherapy (Doxorubicin/Ifosfamide) and then radiotherapy. This patient had several thrombocytopenic episodes and encephalopathy due to ifosfamide. Hence, the patient was diagnosed with B ALL with a very poor prognosis secondary to the sarcoma treatment. Induction chemotherapy with cytarabine and anthracycline was started, but only one cycle of chemotherapy was achieved due to general deterioration, leading to an admission to intensive care. The patient deteriorated rapidly and passed away.

Patient’s molecular features

Bone marrow conventional cytogenetic analysis showed a complex hyperdiploid karyotype with multiple structural and numerical abnormalities including 2p11 deletion and duplication of the inverted chromosome 11. Based on R-banding analysis, the karyotype was interpreted as 63, XX, +del(2)(p11),+5,+6,+7,+8,+9,+9,+10,+10,+11,+12,+13,+14,+15,+19,+20,+22[22]/46, XX [4] (Fig. 1c). FISH analysis was performed on bone marrow aspirate (Fig. 1a) using a LSI KMT2A break apart probe (Vysis, Abbott Molecular Inc.). It revealed the presence of one normal chromosome and two inverted 11 chromosomes, each carrying a KMT2A gene rearrangement with an unknown fusion partner, in four metaphases and in 58/100 nuclei (Fig. 1b).

Fig. 1
figure 1

Cytogenetic analysis at initial diagnosis. Ish inv. (11) (q21, 5’MLL+)(q23, 3’MLL+) ×2 [4].nucish (Mllx3, 5’MLLsep3’MLLx2) [58/100]. FISH study using a LSI KMT2A break apart probe (Vysis, Abbott Molecular Inc.): (a) an interphase cell showing one non-rearranged orange/green signal fusion and two centomeric (red) and two telomeric (green) separate signals, (b) a metaphase cell showing a normal chromosome 11 and two inverted chromosome 11, (c) karyotype from bone marrow aspirate showing complex chromosome abnormalities

Full size image

RNA sequencing (RNA-Seq) has been used to identify gene fusions on RNA extracted from the patient’s blood. Venous blood (10 mL) was collected in EDTA tubes from the patient and PBMC were isolated by erythrocytes lysis (Buffer EL, Qiagen). Total RNA was extracted from 1 × 107 cells using Trizol Reagent (life technologies). For paired-end RNA-Seq, we have used the TruSeq RNA Library Prep Kits followed by paired-end cluster generation and sequencing using the 2 × 150 NextSeq 500 high output kit and the Illumina NextSeq 500 sequencing platform. The first variant (Var 1) corresponded to the fusion of the exon 7 of KMT2A and the exon 2 of MAML2 as described previously [3]. For the second variant (Var 2), the data revealed that the fusion point was within the exon 9 of KMT2A gene and within the intron 1–2 of MAML2 gene. To validate these RNA-Seq results, we have performed RT-PCR with specific primers localized on exon 7 or exon 8 of KMT2A gene and intron 1–2 or exon 2 of MAML2 gene and Sanger sequencing (Fig. 2). The results had confirmed the presence of the two variants but had showed unexpectedly a new variant (Var 3). This fusion gene corresponded to the rearrangement juxtaposing exon 8 and exon 2 of the KMT2A and MAML2 genes respectively and was detected with a higher expression level than the other variants (Fig. 3).

Fig. 2
figure 2

RT-PCR analysis of KMT2A-MAML2 fusion gene. RT-PCR for the KMT2A-MAML2 fusion was performed from patient’s RNA. 1 μg total RNA were reverse-transcribed into cDNA in a 20 μl total volume using random hexamer primers according to EAC protocol (Gabert et al.,2003). PCR reactions contained 1× PCR buffer (Applied Biosystems), 20 nmol dNTPs (Applied Biosystems), 300 nM primer of each primer, 1.25 units of AmpliTaq Gold polymerase (Applied Biosystems), and 5 μl of cDNA in a 50-μl reaction volume. Specific primers were used and were localized on exon 8 of KMT2A (5′-GTCCAGAGCAGAGCAAACAG-3′) and intron 1–2 of MAML2 (5′-TCCCATCTCCAAGTCTCAGC-3′) (Lane 1), on exon 8 of KMT2A and exon 2 of MAML2 (5′-GAGTCTCTCCTGGCTCCTTC-3′) (Lane 2) and on exon 7 of KMT2A (5′-ATCCTGCCCCAAAGAAAAGC-3′) and exon 2 of MAML2 (Lane 3). PCR products were analyzed with Agilent DNA 1000 kit using the 2100 bioanalyzer

Full size image
Fig. 3
figure 3

KMT2A and MAML2 fusion transcripts. The KMT2A exon 7 and exon 8 were fused in-frame with MAML2 exon 2 corresponding to variant 1 (VAR1) and variant 3 (VAR3) respectively. The variant 2 (VAR2) was the resulting of a breakpoint within KMT2A exon 9 and within MAML2 intron 1–2

Full size image

Discussion and conclusions

Here, we reported the first case of a B ALL in a 47 years old woman bearing a KMT2A-MAML2 rearrangement detected by NGS RNA analysis. As a result, KMT2A-MAML2 rearrangement can be found in several types of acute leukemia. Moreover, we have detected two new variants of this fusion transcript based on exon fusion analysis. In fact, RNA sequencing juxtaposing exons 9 of the KMT2A to intron 1–2 of the MAML2 gene, and exon 8 of the KMT2A to exon 2 of the MAML2 gene, have not previously been reported.

Three hypotheses may explain these molecular findings. Either three different breakpoints exist, or/and it is the result of alternative splicing. Indeed, mRNA precursors could lead to the production, via alternative splicing, of a fusion transcript exon 8 of the KMT2A gene to exon 2 of the MAML2 gene and exon 7 of the KMT2A gene to exon 2 of the MAML2 gene. Alternative splicing are mechanisms invoked to play an important role in influencing the mutation effect [8]. The possibility of exons’ skipping or inclusion are almost unlimited, thus, any gene could potentially lead to tens or hundreds of distinctive transcripts. Moreover, we ruled out the possibility of a pre-mRNA by RNA sequencing and PCR. In fact, pre-mRNA contains both introns and exons, and requires splicing of introns to produce the final mRNA molecule containing only exons. In our case, after RNA sequencing and PCR only intron 1–2 of the MAML2 gene persists and this variant was found in significant quantity. Each of the putative fusion transcripts was then translated into predicted amino acid sequences and each of the putative fusion proteins characterized. Two of the three variants (Var 2 and Var 3) contained in-frame stops and are unlikely to encode functional proteins. For the third variant (Var 1), the transcript contained the boundary exons and generated putative fusion transcript.

Furthermore, this case highlights one of the limitations of NGS. In fact, two different KMT2A-MAML2 gene fusions were found by NGS technology, but the major fusion variant, juxtaposing exons 8 of the KMT2A gene and exon 2 of the MAML2 gene, was not detected. NGS methods have the ability to detect a wide array of mutation types, which is mostly limited by the need to know all potential off-target capture regions and to properly process the raw data generated by the sequencer [9]. We had to use Sanger sequencing with specific primers to detect this third variant of KMT2A-MAML2 gene fusion, which was clearly demonstrated on the PCR gel (Fig. 2).

Concerning the development of secondary leukaemia following chemo-radiotherapy for sarcoma, it is known that chemo-radiotherapy administered for a primary cancer will increase the risk of secondary neoplasms [10]. Alkylating agents have been reported as affecting the radiation therapy-associated risk for secondary malignancies [11]. However, in this case, the speed of development of a secondary leukemia is unusual. Alkylating agent/radiotherapy-related leukemia has a mean latency period of 5–7 years, are often preceded by a myelodysplastic phase and are frequently associated with complex karyotypes, whereas acute secondary leukaemia after treatment with DNA topoisomerase II inhibitors has a relatively short latency period (several months to 3 years) and is associated with chromosomal translocation involving 11q23, respectively the KMT2A gene [12, 13].

To conclude, clinical outcomes of KMT2A-associated leukemias are often unfavorable and identification of a KMT2A partner could lead to new therapeutic strategies. Moreover, bringing to light several variants of a fusion transcript, such as the KMT2A-MAML2 gene, is important in the follow-up of residual disease.