Cryptic t(6;11) KMT2A rearrangement in a pediatric acute myeloid leukemia patient detected by next-generation sequencing and dual-fusion FISH analysis

To the Editor: KMT2A-rearranged (KMT2A-R) acute myeloid leukemia (AML) is an aggressive leukemia, and prognosis in children depends on the specific KMT2A-fusion partner, with some requiring an allogeneic hematopoietic stem cell transplant (HSCT) for cure.1–4 Karyotype and breakapart fluorescence in situ hybridization (FISH) are typically used to detect KMT2A-R, but cryptic translocations may result in false negatives, thereby affecting risk stratification and treatment decisions. Next-generation sequencing (NGS) can better detect such cryptic translocations and is nowbeing incorporated by some into the diagnostic workup of pediatric AML patients as standard of care.4,5 Here, we report the case of a 20-month-old male who presented with fatigue, bruising, and bleeding. Complete blood count demonstrated pancytopenia with 2% circulating blasts. Bone marrow (BM) evaluation showed approximately 90% blasts (Figure 1A) with monocytic differentiation by flow cytometry, confirming the diagnosis of AML (Figure 1B,C). G-banded chromosome analysis performed on BM (band resolution 450) showed a complex karyotype: 44,X,Y,der(12)t(12;18)(p11.2;q12),-18[5]/46,XY,add(20)(p13)[4]/82∼85,X X,-Y,-Y,+1,del(1)(p21),del(1)(p22p36.1),-3,-9,-13,-14[cp3]/46,XY[9] (Figure 1D–F). A FISH panel was negative for KMT2A-R (break-apart probe) as well as del(5), del(7), t(8;21), and inv(16) (Figure 1G). However, NGS performed on BM (FoundationOne Heme panel) showed a t(6;11) KMT2A–AFDN fusion (AFDN formerly known as AF6 orMLLT4), confirmed both by DNA and RNA sequencing. Additionally, variants in FLT3 (p.Val592Asp), TP53 (p.Glu326Ter), KRAS (p.Gly13Asp), and NRAS (p.Gly12Cys) were identified. Given the discrepant result for KMT2A-R, dual-fusion FISH for KMT2A–AFDN was performed (Mayo Clinic Laboratories). KMT2A–AFDN fusion was present in 90.2% cells, thus confirming the presence of a cryptic translocation (Figure 1H). The patient received induction I with ADE+GO. End of induction I BM was negative for minimal residual disease. The presence of the t(6;11) KMT2A-R classified him as high risk, requiring HSCT for cure. This would have been undetected had NGS not been performed. He continued to induction II with Mito/AraC with a plan to proceed to HSCT afterwards. Unfortunately, he had an early relapsewith 20%BM blasts and extramedullary disease making him unable to proceed to HSCT. Multiple salvage regimens were attempted; however, he eventually succumbed to his disease. Cryptic translocations are translocations of very small amounts of DNA,which are too small to visualize by karyotyping and also too small to bind a FISHprobe, which is very large by comparison.While PCR can detect cryptic translocations, for example, PML-RARA,6–8 its sensitivity is limited when fusions can occur over a wide breakpoint region. By contrast, NGS can sequence longer ranges of the genome, and only one of the gene partners needs to be in the target region while the other can be determined bioinformatically.9–12 The ability of NGS to detect cryptic translocations and a variety of fusions in a single assay make it a more cost-effective tool. Previously, the presence of a KMT2A-R did not affect risk stratification in pediatric AML. However, a retrospective evaluation of Children’s Oncology Group (COG) phase 3 trial AAML0531 showed that t(6;11)(q27;q23), t(10;11)(p11.2;q23), t(10;11)(p12;q23), and t(11;19)(q23;p13.3) were associated with an extremely inferior prognosis, with a 5-year event-free survival of 13%, 17%, 20%, and 14%, respectively.3,13 Accordingly, the current COG phase 3 AML trial, AAML1831, now classifies all these KMT2A-R as high risk, thus requiring HSCT in first remission to achieve the highest likelihood of cure.5 Therefore, identifying such lesions accurately are paramount to treatment determinations, making NGS critical to ensure that rare cryptic translocations are not missed.4,5