Recharacterization of cancers and other diseases in pathophysiological terms is key to the future of medicine. Considerable progress has been made in the molecular classification of some cancers, such as hematological malignancies (Box 1-1). More recently, the systematic analysis of genomic alterations in a small set of breast and prostate cancers revealed that individual tumors accumulate an average of approximately 90 mutant genes but that only a subset of these contribute to the neoplastic process. The authors identified 189 genes (average of 11 per tumor) that were mutated at
Biomarkers of Hematologic Cancers
The diagnosis of hematologic cancers presents an enormous challenge. The numerous stages of hematopoietic differentiation give rise to many biologically and clinically distinct cancers, most often via acquired genetic alterations. Knowledge of the biology underlying hematological malignancies has greatly increased in recent decades, leading to a much more sophisticated classification system that incorporates not only the traditional morphologic characteristics, but also immunophenotypic, genetic, and clinical features. However, even with this added information, considerable heterogeneity still exists within identified subtypes, with different clinical presentations and outcomes.
Researchers have long sought a classification system based on molecular pathogenesis, and DNA microarrays have been recently used to survey the expression of thousands of genes in parallel. Studies have identified novel disease subtypes and have also uncovered relationships between diseases previously considered to be unrelated. The results show great promise for refining diagnosis and prognosis, predicting response to treatment, and identifying potential targets for novel therapeutic interventions, although much work remains to be done before such tests can be routinely used to aid clinical decisions. For example, further clinical validation in larger cohorts and independent studies are needed, as well as test platform standardization and analytical validation. It is also not yet clear whether whole gene expression patterns are required, or whether a small set of genes will be sufficient to predict prognosis.
SOURCES: Reviewed by Staudt, 2003; Levene et al., 2003; Bullinger, 2005; Bullinger et al., 2005.