Transcriptomics-based multi-omics approach for optimizing risk stratification in acute myeloid leukemia

Acute myeloid leukemia (AML) is a subtype of hematopoietic neoplasm affecting myeloid cells in blood and bone marrow. Molecular subtyping of AML provides in-depth insights into disease pathogenesis, facilitating clinical diagnostics and therapeutic management. While current molecular screening is mainly established on DNA-based approaches, including targeted sequencing of gene panels and polymerase chain reaction (PCR)-based fusion gene detection, transcriptome-based RNA-seq has gained increasing attention for its potential in improving molecular subtyping as well as in facilitating prognostic prediction and therapeutic selection. In this study, we aimed to identify transcriptomic signatures in molecular genotyping and risk stratification in a cohort of 125 patients with AML excluding those harboring <i>PML::RARA</i> fusion genes. We subsequently established a 2-step diagnostic assay based on transcriptomic expression, which can be further explored for predicting the outcome of AML patients with increased confidence and accuracy. Together with the identification of genetic mutations and fusion genes, transcriptome-based stratification might significantly enhance the ability of outcome prediction. In summary, we developed a transcriptomics-based multi-omics prediction model for improved risk stratification of patients with AML.