A team of University of Florida Health researchers has developed a digital tool that uses artificial intelligence to accelerate the diagnosis of acute leukemias. The research was published in Nature Communications.
Acute myeloid leukemia, or AML, is a blood cancer originating in bone marrow and is often fatal. It’s one of the most common leukemias in adults and can progress quickly.
A team led by Jatinder Lamba, Ph.D., associate dean for research and graduate education and a professor of pharmacotherapy and translational research in the UF College of Pharmacy, developed the Acute Leukemia Methylome Atlas, or ALMA, by mapping specific tags in DNA referred to as methylation patterns across 3,300 leukemia samples. The result is a tool that can match patients to 27 leukemia subtypes as defined by the World Health Organization.
The atlas is an open-access reference web tool that lets any new patient’s DNA methylation results be placed among thousands of previously diagnosed cases, allowing for rapid diagnosis of the leukemia subtype. Using an algorithm, the atlas plots each patient as a point. Points that sit close together share similar methylation patterns and often the same underlying leukemia subtype or risk category. The atlas allows any patient sample to be compared to thousands of other cases, instantly showing how their cancer maps to a specific subtype.
Researchers have developed two AI-powered tools that can help clinicians predict how a patient’s disease might progress and their chances of survival. One tool estimates the likelihood of surviving five years based on genetic markers, while the other uses a smaller set of markers that could make quick lab tests possible to guide treatment decisions.
“Currently, clinicians often wait weeks for diagnostic lab results, but with our pipeline, this wait time can be reduced to 2-3 days for a preliminary diagnosis. This single-test assay, requiring only a laptop-sized sequencer, can be run in-house, lowering costs, widening access and improving long-term remission globally,” said Lamba, the Frank A. Duckworth Eminent Scholar Chair who also serves as a co-leader of the UF Health Cancer Center’s Cancer Targeting and Therapeutics research program. “By using our prognostic features, a clinician can predict whether the patient has a higher likelihood of having a relapse, allowing the health care team to make decisions about intensifying the patient’s therapy, monitoring the patient more closely or even reassessing the need for transplants.”
Lamba said the research team aims to explore its findings in a pilot clinical trial and to refine the ALMA into a tool for pathologists and clinicians. Future research will involve gathering data on patients with rare subtypes of AML to create an even more complete atlas, she said.
Francisco Marchi, a graduate student in the Lamba Lab, is a co-inventor of this technology and first author of the study. Other co-authors include Vivek Shastri, Ph.D., a research assistant professor of pharmacotherapy and translational research in the UF College of Pharmacy, along with faculty from the UF College of Medicine and the UF Health Cancer Center. UF graduate and undergraduate students also contributed.