Phenotypic correlates of genetic abnormalities in acute and chronic leukemias

The first classification of the acute leukemias in 1976 relied exclusively on the evaluation of cell size, granularity, nuclear shape, cytoplasmic appearance, cytochemical reactions, and dysplastic features of cells surrounding the "leukemic blast." In selected cases, researchers have even succeeded in not only elucidating but also reversing the oncogenic mechanism, such as in acute promyelocytic leukemia (APL). APL has become the paradigm for the ultimate goal of leukemia diagnosis: to be able to tell an oncologist what targeted therapy a patient is a candidate for, based on the detection of a specific genotype. Some antibodies invariably stain all cells from a given lineage but vary markedly in their intensity of staining between normal and malignant cells, suggesting variable antigen densities (e.g., those of CD20 and CD22 on normal versus chronic lymphocytic leukemia [CLL] B lymphocytes); for other antigens, the fraction of cells binding the antibody will contain the diagnostic information (e.g., the percentage of CD34 ⁺ or CD117 ⁺ cells in any acute leukemia). Acute myeloid leukemia (AML), acute lymphoid leukemia (ALL), or chronic myelogenous leukemia (CML) tissues should be cultured for 24 h without mitogens before processing. Samples from CLL require stimulation by B-cell mitogens (B-cell CLL) or T-cell mitogens (T-cell CLL) for cell division to occur. The principle of targeted therapy, as attractive and promising as it is, requires much more understanding of transforming molecular events and perturbed signaling pathways than the simple administration of indiscriminately cytotoxic chemotherapy.