Acute myeloid leukemia (AML) is the most common form of acute leukemia in dogs and the most devastating as far as outcome.1–4 Despite chemotherapy, most dogs with AML die or are euthanized within a few days to weeks of diagnosis, with only a few dogs surviving for longer (several months). Much of our current knowledge of canine AML is restricted to disease diagnosis, including expected clinical signs and laboratory test results. Unfortunately, this information is of little prognostic or therapeutic use, with our treatment options currently limited to traditional lymphoma or “AML-like” protocols, with choice of treatment being per clinician preference versus evidence-based. Canine AML mimics the human counterpart, with respect to clinical and laboratory findings, biologic behavior, and response to therapy. Yet in humans, substantive advances have been made with the advent of genetic testing. Genetic analysis has resulted in the detection of driver mutations, mutations that are targetable with new drugs, and mutations that stratify patients into subcategories of disease with prognostic implications. Indeed, the diagnosis and classification of AML in humans is now reliant on genetic testing. Our lack of knowledge of the basic molecular biology of canine AML is impairing our ability to treat dogs effectively for the disease, provide prognostic information to owners, and improve outcomes. We propose to begin to fill this knowledge gap by performing unbiased whole exome sequencing on 20 well-phenotyped canine AML.

With the aid of our co-investigators at the University of Georgia, Guelph and Kansas and collaborators from specialty clinics, we will enroll 20 dogs with AML into the study. The diagnosis of AML will be based on clinical pathologic data documenting >20% blasts in blood and/or bone marrow, with myeloid lineage confirmed on flow cytometric immunophenotyping and cytochemical staining, as we have described. Genomic DNA will be isolated from neoplastic cells in blood or bone marrow and germline DNA will be harvested from oral mucosa swabs. DNA library preparation and exome capture will be accomplished using a canine-specific kit, with DNA sequencing performed at 200x depth in the Genomics Core in the Biotechnology Institute at Cornell University. Bioinformatics analysis will then be performed to identify single nucleotide variations (SNV), insertions-deletions (indels) and copy number variations (CNV). We will stratify the dogs into specific subtypes of AML (e.g. monocytic, myelomonocytic) and determine if specific gene mutations are linked to these subcategories.