Rapid Epigenomic Classification of Acute Leukemia using Deep Learning and Nanopore Sequencing.
Acute leukemia (AL) is a rapidly progressive blood cancer that requires immediate treatment. AL is biologically heterogeneous with multiple distinct disease categories, many of which have direct impact on prognosis or eligibility for targeted therapies. Thus, effective management of patients presenting with AL requires a precise and timely diagnosis. AL classification is currently built on a multitude of diagnostic assays that must be performed in parallel, including morphologic review of bone marrow biopsies and peripheral blood, immunophenotyping, and genetics analysis. While informative, these tests can require specialized infrastructure and expertise to perform and interpret, and can lead to unnecessary costs, confusion, and delays in care. Moreover, these approaches often lack the specificity to resolve clinically relevant disease states or predictive categories in AL.
To address these challenges, our team is developing a framework to rapidly classify AL in the clinic using DNA methylation profiling and real-time data analysis using a machine learning classifier. Our approach offers several key advantages: Single-assay epigenetic profiling effectively resolves the biological heterogeneity of AL and is a valid surrogate for many conventional diagnostic assays. When coupling our classifier to emerging Nanopore sequencing, clinically actionable results can be generated within hours rather than days. In addition, Nanopore sequencing is affordable and easy to implement, making it suitable for state-of-the-art clinical laboratories but also remote settings without ready access to hematopathology or molecular diagnostics services. We believe that our framework has great potential for application in clinical routines and in basic research and provides a foundation for future developments in machine learning-assisted diagnostics for AL and other cancers.