A Systematic, Evidence-Based Workflow for Classifying KMT2A Fusions in Acute Myeloid Leukemia

KMT2A fusions are a critical oncogenic driver in 5% to 10% of patients with acute myeloid leukemia (AML) and are associated with poor prognosis. Currently, there are no published somatic guidelines for fusions in AML, and developing methods to accurately classify fusions, especially those involving KMT2A, is essential for patient care. Therefore, the Laboratory for Personalized Molecular Medicine (LabPMM) KMT2A Fusions Workflow was developed utilizing the framework of the somatic guidelines by Horak et al, where classification of oncogenicity is based on points awarded for varying types of evidence. Fusions previously detected by LabPMM's CAP/CLIA-certified MyAML and MyMRD gene panels were used to test this workflow. A total of 100 KMT2A fusions were reassessed, and 97 of these had a breakpoint in the major breakpoint cluster region. There were 20 distinct partner genes for KMT2A, and the most common partners were MLLT3, ELL, AFDN, MLLT10, and AFF1. Five KMT2A fusions had a novel partner (MYB, RC3H1, SNAPC3, STPG1, and HPSE2). Breakpoints in the partner genes were assessed to better understand their potential role in driving leukemogenesis. Of the 100 fusions reassessed, 9 had a classification change. This LabPMM KMT2A Fusions Workflow provides a points-based system for curation that allows for standardization and clarity both within and among genetic diagnostic laboratories reporting on KMT2A fusions in AML.