Abstract
Acute myeloid leukemia (AML) remains a major therapeutic challenge despite remarkable advances in genomic characterization. Although molecular profiling has transformed disease classification and risk stratification, treatment for most patients continues to rely on cytotoxic chemotherapy regimens developed more than 50 years ago. Numerous attempts to improve outcomes through treatment intensification, alternative chemotherapy approaches, and targeted agents have produced incremental benefits for selected patient populations, but have failed to fundamentally alter outcomes for many high-risk AML subtypes. A major obstacle to the broader application of immunotherapy in AML has been the absence of targets that clearly distinguish leukemic cells from normal hematopoietic tissues. Consequently, most immune-based therapies have focused on AML-associated antigens such as CD33, CD123, and CLEC12A, resulting in a narrow therapeutic window and significant hematopoietic toxicity.
Recent advances in large-scale transcriptomic profiling, integrated genomic analyses, and immunopeptidomics have enabled systematic identification of AML-restricted biomarkers linked to leukemia-defining oncogenic programs. Rather than seeking a universal AML antigen, these approaches have uncovered subtype-specific targets, including FOLR1 in CBFA2T3::GLIS2 AML, CLEC2A in KMT2A-rearranged AML, mesothelin or CD7 in selected high-risk AML subsets, and intracellular targets such as WT1, PRAME, NPM1 neoantigens, and fusion-derived peptides. These discoveries have created opportunities for immunotherapeutic approaches capable of selectively targeting leukemic cells, while preserving normal hematopoiesis.
This Spotlight Review discusses the evolution of AML biomarker discovery from prognostic classification to therapeutic target identification and highlights emerging AML-restricted biomarkers that may enable a new generation of biologically-precise immunotherapies for molecularly defined AML subsets.
Figures & Tables
Article Information

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.