63. Comparative analysis of testing methods used for the detection of internal tandem duplications in the KMT2A/MLL gene

Aly Abdelkareem

José-Mario Capo-Chichi

Dr Capo-Chichi is a lab director at the Genome Diagnostics Laboratory at University Health Network (UHN) in Toronto and an assistant professor in the Department of Laboratory Medicine and Pathobiology at the University of Toronto. Dr Capo-Chichi is also a clinician investigator at the Princess Margaret Cancer Centre Research Institute with a primary focus on the genomic bases of acquired and inherited myeloid malignancies. Dr Capo-Chichi’s interest include the improvement of traditional genomic approaches and the development of novel clinical tests for diagnosis and prognostication of cancers, using high-throughput genotyping strategies. 


José-Mario Capo-Chichia, Andrew Setoa, Olivia Kinga, Shabnam Salehi-Rada, Ana Baptistaa, Grenier Sylviea, Bev Nwachukwua, Anne Tierensa, Andrea Arrudab, Mark Mindenb, Adam Smitha

aLaboratory Medicine Program, University Health Network, Toronto, Ontario, CA; bPrincess Margaret Cancer Centre, University of Toronto, Toronto, Ontario, CA

The Lysine (K)-Specific Methyltransferase 2A (KMT2A) gene (formerly MLL) that resides on human chromosome 11q23 is frequently rearranged in de novo and therapy related leukemia. KMT2A translocations with various partner genes are often easily detected by classic molecular and cytogenetic approaches. However, the detection of small intragenic insertions, the partial tandem duplications of the KMT2A gene (KMT2A-PTD), is challenging. Additional co-occurring alterations, such as Trisomy 11 which alerts one to the PTD, may further hamper the identification of KMT2A-PTD.

KMT2A-PTDs occur in 3-10% of adult AML and the presence of a KMT2A-PTD is associated with a poorer AML outcome. As such, accurate detection of KMT2A-PTDs in AML routine clinical diagnostic workup is important. Here we assessed the ability of three orthogonal technologies for KMT2A-PTD analysis :1) an hybridization-capture based Next Generation Sequencing targeting the KMT2A gene region, 2) an MLPA based assay covering exons 4 and 36 of KMT2A and 3) Optical Genome Mapping.

We analyzed tumor samples of ten AML patients with KMT2A-PTD and evaluated the relative performance, strengths and weaknesses of these technologies. KMT2A-PTDs are in-frame duplications that typically encompass exons 3 through 9. However, our results revealed the presence of unsuspected variants of KMT2A. Specifically, copy number based approaches, in the absence of structural data, can lead to false positive and negative calls due to genomic alterations, such as unbalanced translocations, involving KMT2A. Therefore, our data suggest that a combined structural and copy number approach are required to accurately detect KMT2A PTDs.