The concepts of cancer stem cells in general, and leukemia stem cells in particular, have themselves undergone a significant evolution over the last decade. Initially, clonal evolution of stem cells was thought to represent a fundamentally linear process1 in which an initiating genetic aberration was complemented by one or more “driver” mutations, including copy number alterations (CNAs). This model suggested that if such stem cells could be identified and their genetic abnormalities characterized, specific targeted therapies could be designed to eradicate the initiating clones, leading to disease regression. However, over the last several years, multiple groups have described findings that are difficult to reconcile with the linear clonal evolutionary model of cancer (or leukemia) stem cells. These findings include the genetic and phenotypic heterogeneity of such cells, as well as their disparate growth patterns in immunocompromised mice.
Two recent studies involving acute lymphoblastic leukemia, results of which have recently been reported in Nature, highlight these issues and place them in an essentially new perspective. A group led by Dr. Mel Greaves of the Institute of Cancer Research, Sutton, United Kingdom, examined a series of pediatric acute lymphoblastic leukemia cases characterized by the ETV6/RUNX1 fusion gene in relation to the presence of CNAs. Using techniques capable of single cell analysis, these investigators observed highly heterogeneous and diverse clonal architectures, most consistent with a branching, non-linear evolutionary history of leukemogenesis. Significantly, CNAs occurred in various subclones in no particular order and reiteratively at various stages of the disease. Leukemia cells propagating in immunodeficient mice also showed heterogeneous genetic alterations and proliferative capacities. In a parallel study by Dr. John Dick and colleagues from the Ontario Cancer Institute that investigated CNAs in patients with BCR/ABL1 lymphoblastic leukemia, the authors observed very similar findings most consistent with a non-linear, branching, multi-clonal model of leukemogenesis.2 Of note, patients whose cells exhibited a deletion of CDKN2A/B had a particularly poor prognosis in that study.
In Brief
The notion that putative leukemia stem cells are considerably more complex in their genetic makeup and biologic behavior than previously appreciated is sobering, yet it offers a theoretical foundation for future attempts to develop effective stem-cell-directed therapies. At one level, the concept of a heterogeneous, branching, non-linear evolution of leukemia stem cells could help to reconcile conflicting reports in the literature attempting to explain their origin and predict their behavior. At a second level, these insights could have important translational implications for the leukemia stem cell field. Although it would be comforting to think of leukemia-initiating cells in a monolithic way, it is becoming apparent that such a view represents a gross oversimplification and that individualized therapy may be necessary to eradicate multiple stem cell clones. This is undeniably a daunting task, but the present findings provide a rational basis for beginning this important quest.
References
Competing Interests
Dr. Grant indicated no relevant conflicts of interest.
