Introduction

Acute myeloid leukemia (AML) is an oligoclonal malignancy characterized by several differences from normal hematopoiesis, including blockade of normal differentiation, enhanced self-renewal, increased proliferation, impaired apoptosis, dissemination, and genomic instability. Recent advances in our understanding of the pathogenesis of AML have involved elucidation of the role of aberrant chromosomal rearrangement, amplifications, deletions, and point mutations, and aberrant regulation of gene expression, governed in part by changes in chromatin. AML can now be subclassified based upon gene mutations, gene expression, miRNA expression, and DNA methylation states, and many of the phenotypic properties of AML and alterations in gene expression can be mapped onto underlying genetic lesions. Within several years, a near complete categorization of AML will be achieved, and a variety of new therapeutic targets will be identified. Remaining challenges will be to understand the molecular mechanisms linking genetic and epigenetic changes to leukemia cell growth and the translation of these findings into specific therapies that may target mutant proteins or deregulated gene networks.

The evolving modes of classification of acute myeloid leukemia

The French-American-British (FAB) system classified acute myeloid leukemia (AML) by analogy to normal myelopoiesis, but yielded relatively poor information in that only a few subtypes, such as acute promyelocytic leukemia (APL), could be distinguished as having a distinct prognosis. Over the past two decades, conventional cytogenetics, flow cytometry, fluorescence in situ hybridization, DNA sequencing, and PCR have led to the definition of more precise subsets of AML with distinct prognoses. This was reflected in the 1999 WHO system that includes cytogenetic and molecular anomalies (1). The revolution in genomic technology will soon lead to a reassessment of AML and the definition of even smaller prognostic subsets.